Spatial Variation and Risk Assessment of Trace Metals in Water And

Spatial variation and risk assessment of trace metals in water and sediment of the Mekong Delta Emilie Strady, Quoc Dinh, Julien Némery, Thanh Nho Nguyen, Stéphane Guédron, Nhu Sang Nguyen, Hervé Denis, Phuoc Dan Nguyen

To cite this version:

Emilie Strady, Quoc Dinh, Julien Némery, Thanh Nho Nguyen, Stéphane Guédron, et al.. Spatial variation and risk assessment of trace metals in water and sediment of the Mekong Delta. Chemosphere, Elsevier, 2017, 179, pp.367-378. 10.1016/j.chemosphere.2017.03.105. hal-02357361

HAL Id: hal-02357361 https://hal.archives-ouvertes.fr/hal-02357361 Submitted on 13 Nov 2019

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Spatial variation and risk assessment of trace metals in water and sediment of the Mekong Delta

Emilie Strady a, b, *, Quoc Tuc Dinh b, c, Julien Ne´mery a, b, Thanh Nho Nguyen d, Ste´phane Gue´dron e, Nhu Sang Nguyen c, Herve´ Denis a, Phuoc Dan Nguyen c a Univ. Grenoble Alpes, CNRS, IRD, Grenoble INP, IGE, F-38000 Grenoble, France b CARE-HCMUT, Ho Chi Minh City, Viet Nam c Faculty of Environment, HCMUT, Ho Chi Minh City, Viet Nam d Faculty of Chemistry, University of Sciences, Ho Chi Minh City, Viet Nam e Univ. Grenoble Alpes, CNRS, IRD, ISTerre, F-38000 Grenoble, France h i g h l i g h t s

● Trace metal transportation are controlled by SPM concentrations. ● Dissolved trace metal distribution vary in the salinity gradient. ● Ecotoxicological indexes in surface sediments show low to medium contamination. a r t i c l e i n f o a b s t r a c t

Article history: The Mekong Delta, is home to 17 million inhabitants and faces numerous challenges relating to climate Received 29 November 2 0 1 6 change, environmental degradation and water issues. In this study, we assess trace metals concentrations Received in revised form (Al, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Mo, Cd, Hg, Pb) in the water, suspended particulate matter and surface 23 March 2017 sediments of the Tien River, the Northern branch of the Mekong Delta, during both dry and rainy seasons. Accepted 26 March 2017 Available online 27 March 2017 Metal concentrations in the dissolved and suspended particle phases remain in the low concentration range of the main Asian Tropical River. During transportation in the riverine part, we evidenced that V, Cr, Co, As and Pb are dominant in the particulate phase while Mo, Ni and Cu dominate in the dissolved fraction. In the salinity gradient, dissolved U, V, Mo exhibit conservative behaviour while Ni, Cu, As, Co Keywords: and Cd showed additive behaviour suggesting desorption processes. In the surface sediment, metal Mekong Delta concentrations are controlled by the particle-size, POC contents and Fe, Al and Mn e oxy(hydr)oxides. Trace metals Calculated Enrichment Factor and Geoaccumulation Index evidenced As enrichment while the calculated Ecotoxicological index mean effect range median quotients evidenced a low to medium ecotoxicological potential effects range Tropical river in the surface sediments. Estuary

1. Introduction 2008; de Souza Machado et al., 2016; Garnier et al., 2010). Because of their persistence, toxicity and ability to accumulate in Deltas and estuaries play a major role in material and element organisms, trace metals are major pollutants and are considered a transport from river to ocean and on biogeochemical cycles. The high priority (i.e. European Water Framework Directive various physical and chemical gradients that occur in this mixing (Anonymous, 2000), US-EPA (40 CFR Part 423, Appendix A). Trace environment affect the partitioning, mobility and reactivity of el- metals in aquatic environments originate mostly from natural ements such as nutrients and trace metals (e.g. Du Laing et al., erosion and soil leaching. They are also released by human activity such as industrial, domestic, urban and agricultural practices. Suspended particles are a key factor in contaminant transport from the continent to the ocean, making their quality assessment a major * Corresponding author.Univ. Grenoble Alpes, CNRS, IRD, Grenoble INP, IGE, F- scientific concern (e.g. Apitz and Power, 2002). Understanding the 3 8 0 0 0 Grenoble, France. processes controlling elemental transport and reactivity between E-mail address: [email protected] (E. Strady). the dissolved and particulate (both suspended and deposited sediments) phases in rivers and more particularly in estuaries is thus crucial to assess their impact in terms of contamination and ecotoxicological risks. The Mekong River is the longest River in Southeast Asia with a total length of 4800 km and a drainage basin of 795,000 km2. It originates in the Tibetan Plateau of western China and flows southward through China, Myanmar, Laos, Thailand, and Cambodia, before entering Vietnam through the Delta and then discharging into the South China Sea. The Mekong Delta is one of the world’s largest deltas (39 0 00 km2) and has a population of around 17 million people (GSO, 2016). It is the principal area of rice production in Vietnam and its economy is largely reliant on agriculture (namely rice and fruit) and aquaculture (mainly shrimp and cat- fish). Despite recent intensification of agricultural and aquaculture activity as well as rapid urban growth, the Mekong Delta remains one of the poorest regions in Vietnam (Renaud and Künzer, 2012). The delta’s natural and social systems face numerous challenges related to climate change, environmental degradation and water issues such as flooding, water pollution and access to water (Xue et al., 2011; Renaud and Künzer, 2012). Hydroelectric dams on the Mekong River and its tributaries have severely impacted the aquatic ecosystem’s biodiversity (Campbell, 2012). Sediment sup- ply to the delta and the ocean is also impacted, and estimates Fig. 1. Sampling location map. suggest that it retains about 32e41 million tons of sediment per year (Kummu et al., 2010). Indeed, before dam construction, the marine and fluvial origin that were rapidly deposited beginning Mekong was one of the 10 largest sediment suppliers to the world’s 80 00 year BP (Nguyen et al., 2 0 00 ; Xue et al., 2011). The climate is oceans with an annual sediment flux estimated at around 160 monsoonal humid and tropical, with average temperatures of million tons (Milliman and Meade, 1983; Milliman and Ren, 1995). 27e30 ○C. The rainy season (approximately 80% of the annual Numerous dams are currently under construction or in the plan- rainfall) lasts from May to October. Accordingly, the Mekong River ning stages on the main stream and in the tributaries, increasing discharge reaches a minimum in AprileMay and a maximum in the vulnerability of the Mekong River. The Mekong Delta river SeptembereOctober in the lower Mekong (Xue et al., 2012) and the quality is also threatened by the development of intensive agri- same is true for the Bassac River where the river discharge fluctu- cultural and aquaculture activity and the release of pesticides (Toan ates from 200 m3 s—1 to 7000 m3 s—1 respectively (Loisel et al., et al., 2013), antibiotics (Giang et al., 2015), nutrients and trace 2014). metals (Wilbers et al., 2014). Information on trace metal risk assessment in the Mekong Delta’s waters and surface sediments is not readily available (e.g. 2.2. Sampling and handling Cenci and Martin, 2004; Noh et al., 2013; Wilbers et al., 2014) and studies that are available have mainly focused on arsenic ground- Two snapshot campaigns were conducted along the Tien River water contamination issues and consequences to population health during dry and wet seasons, in March and October 2013, respec- (Berg et al., 2007; Buschmann et al., 2008; Hoang et al., 2010). Thus, tively. At each site (Table 1a), temperature, pH, dissolved oxygen, considering the environmental challenges that are facing the conductivity and salinity were immediately measured in situ using Mekong Delta, the purpose of this study is to evaluate trace element a multi parameter probe (WTW 3420®). Then, water was sampled contamination (Al, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Mo, Cd, Hg, Pb, U) at 10e50 cm below surface using a Niskin non-metallic water of the Tien River, Mekong Delta, Vietnam. Two snapshot campaigns sampling bottle (General Oceanic®) and stored in 5L PE bottle. All were performed along the Tien River during the contrasted seasons filtered and unfiltered samples were stored in a cooler (~4 ○C) until (dry and rainy seasons). The objectives were (i) to get an inventory being brought back to the laboratory. A first filtration was per- of dissolved and particulate polymetallic concentrations which formed on glass microfiber filter (0.7 mm GF/F Whatman®, pre- affect the quality of the river, (ii) to identify the fate of metal dis- weighed and preheated at 500 ○C). The filtrate was stored in a tribution in the water column and sediment and the factors con- 60 ml bottle and kept at —18 ○C for analyses of dissolved nutrient trolling their partitioning, and (iii) to provide a risk assessment of while filters, used for determination of suspended particulate surface sediments based on geochemical and ecotoxicological matter (SPM) and particulate organic carbon (POC) concentrations, indexes. were dried at 50 ○C, then weighed and stored at room temperature. A second filtration was realized on pre-weighed PTFE filters (0.20 2. Material and methods mm Omnipore®) for analysis of particulate and dissolved trace metals. After filtration, the filtrate was acidified (Normapur HNO3 2.1. The Tien River, Mekong Delta, Vietnam 2% v/v) and stored in a 30 ml acid pre-cleaned PP bottle (Normapur ○ HNO3 10% v/v) at 4 C while the filters were kept in sterile plastic The Mekong Delta begins in Phnom Penh, Cambodia, where the petri dishes at —18 ○C, then freeze-dried, weighed and stored in the river divides into its two main branches, the Mekong and the plastic petri dishes. To prevent contamination during collection and Bassac, which are respectively subdivided into six and three handling, all equipment was cleaned and we wrapped each sub- branches, the Tien River being the northern branch of the Delta sample individually in two polyethylene bags. Surface sediments (Fig. 1, 150 km long, 450e2250 m width, up to 10 m depth). The were collected at each site using a Shipek sediment grab sampler, Mekong Delta is composed of Holocene alluvial sediments of immediately stored in PE bags at —18 ○C and, freeze-dried. Table 1a Location of sampling site: GPS coordinates and main characteristics.

Sampling site Coordinates City Activity

MD-1 N10○ 24051.000 E105 ○ 39011.800 Cao Lanh City, 1 1 0 0 0 0 inhabitants agriculture: rice, fruit; industrial activity MD-2 N10○ 17055.700 E105 ○ 46006.400 Sa Dec City, 1 6 0 0 0 0 inhabitants agriculture: fruit MD-3 N10○ 16038.300 E105 ○ 54024.500 My Thuan City, 1 6 0 0 0 0 inhabitants agriculture: fruit; industrial activity MD-4 N10○ 19009.200 E106 ○ 01012.300 Cai Be City, 2 6 0 0 0 0 inhabitants agriculture: fruit MD-5 N10○ 18057.000 E106 ○ 12002.000 Rural area agriculture: rice, fruit MD-6 N10○ 20053.400 E106 ○ 21002.600 My Tho City, 2 3 0 0 0 0 inhabitants agriculture: fruit; industrial activity MD-7 N10○ 18035.800 E106 ○ 30017.900 Rural area agriculture: rice, fruit; aquaculture: ﬁsh MD-8 N10○ 17011.600 E106 ○ 34056.100 Rural area agriculture: rice, fruit; aquaculture: shrimp MD-9 N10○ 17019.300 E106 ○ 41039.000 Rural area agriculture: rice, fruit; aquaculture: shrimp MD-10 N10○ 15036.800 E106 ○ 45019.000 Rural area agriculture: rice, fruit; aquaculture: shrimp

2.3. Laboratory analyses 2.3.4. Grain size distribution analyses Grain size distribution was measured on bulk sub-samples using 2.3.1. Trace metal analyses a laser diffraction sizer after a 3 min ultrasonic agitation (Malvern Particulate trace metals on PTFE filters (MetalSPM) and repre- Mastersizer 2000, IGE-OSUG Laboratory, Grenoble, France). sentative sub-samples of surface sediments (MetalSED; 100 mg of dried, powdered and homogenized material) were analyzed by a 2.4. Geochemical and ecotoxicological indexes microwave total extractable acid digestion (NovaWave SCP Sci- ® ences , IGE-OSUG Laboratory, Grenoble, France; HCl, HNO3, HF; 2.4.1. Enrichment factor ® Trace Metal grade Fisher ). The protocol is based on the digestion Assessment of the metal enrichment degree in sediments was method USEPA 3052 and is fully described in Strady et al. (2017). undertaken using the enrichment factor (EF). The index allows Trace metal concentrations (V, Cr, Co, Ni, Cu, Zn, As, Mo, Cd, Pb, U) differentiating natural geochemical background to anthropogenic were measured by ICP-MS (Elan DRC II Perkin Elmer, TERA Envi- inputs (Zhang and Liu, 2002). The enrichment factor is defined as ronment Laboratory Fuveau France) while Al, Mn and Fe concen- the ratio of Al normalized metal concentrations in sediments over trations were measured by ICP-AES (Varian 720 ES, ISTerre-OSUG Al normalized ratio in a geochemical background reference. Laboratory, Grenoble, France) using external calibration for both Normalization to Al is used to compensate natural variability due to cases. Total Hg concentrations in sediments THgSED were deter- grain-size variations (e.g. mineral composition) and to detect any mined by atomic absorption spectrophotometry after dry miner- anthropogenic metal contributions (Loring, 1991; Chapman and alization and gold amalgamation by using an automatic mercury Wang, 2001). Zhang and Liu (2002) defined that an EF value be- analyzer (Altec, Model AMA 254 ISTerre-OSUG Laboratory, Gre- tween 0.5 and 1.5 suggests natural weathering processes while a ´ noble, France) (Gue dron et al., 2009; Strady et al., 2017). MetalF value of EF > 1.5 suggests trace metals being delivered from non- concentrations were directly measured by ICP-MS (Elan DRC II crustal materials corresponding to enrichment from anthropo- Perkin Elmer) using external calibration. When the sample salinity genic sources. More precisely, Birth (2003) defined that 1.5 < EF < 3 was up to 1, dissolved trace metals were determined by Kinetic is minor enrichment, 3 < EF < 5 is moderate enrichment, 5 < EF < 10 Energy Discrimination -Argon Gas Dilution (KED-AGD mode) with is moderately severe enrichment, 10 < EF < 25 is severe enrich- the Thermo Scientific iCAPQ ICP-MS (Plateforme AETE- ment, 25 < EF < 50 is very severe enrichment, and finally that EF HydroSciences/OSU OREME, Montpellier, France) using an added > 50 is extremely severe enrichment. on-line internal solution (Sc, Ge, In and Bi) to correct signal drifts. The analytical quality assurance of particulate measurements was 2.4.2. Geoaccumulation Index (Igeo) fi assured by analyzing certi ed reference materials. Accuracy was V: The degree of metal enrichment was also assessed using the 6%; Cr: —6%; Co: 7%; Ni: —2%; Cu: —6%; Zn: 30%; As: 1%; Mo: —3%, Geoaccumulation Index (Igeo, Müller, 1979) originally defined for — — ¼ Cd: 13%, Pb: 8%, Hg: 9% for GBW-07323 (n 5) and V: 1%; metal concentrations in the <2 mm fraction with respect to the — — — — — — Cr: 6%; Co: 7%; Ni: 13%; Cu: 9%; Zn: 29%; As: 5%; Mo: 3%, background value Bn as a ‘pre-civilisation’ value for the study area. — ¼ Cd: 2%, Pb: 9%, Hg: 5% for MESS-3 (n 4). Precision was lower The Igeo calculation is based on the following formula: than 8% for all elements and both GBW-07323 and MESS-3. Cn Igeo ¼ Log 2.3.2. Particulate organic carbon analyses 2 1:5*Bn Acombustion infrared detection technique using a LECOCS-125 analyzer (EPOC Laboratory, Talence, France) with 5% precision was where Cn is the measured concentration in the sediment for metal used to determine POC concentrations. Contents of POC are n, Bn is the background value for the metal n, and factor 1.5 is used expressed as a percentage of the dry weight of SPM or sediment, because of possible variations in background data due to litholog- abbreviated as POC%. ical variations. The geoaccumulation Index includes seven grades: Igeo≤0 Uncontaminated; 0 < Igeo<1 uncontaminated to moder- 2.3.3. Orthophosphates, nitrates and ammonium analyses ately contaminated; 1 < Igeo<2 moderately contaminated; < < < < Orthophosphates (P-PO4) were analyzed using the ascorbic 2 Igeo 3 moderately to strongly contaminated; 3 Igeo 4 acid-molybdate blue method developed by Murphy and Riley strongly contaminated; 4 < Igeo<5 strongly to extremely contam- (1962) prior to spectrophotometric measurements. Ammonium inated; 5 < Igeo extremely contaminated. (N-NH4) was analyzed directly on site using the standard colori- metric method (APHA, 1995) and Hach Lange photolab. Nitrate (N- 2.4.3. Sediment quality guideline ® NO3) was measured by ion chromatography (A-732 Metrohm A sediment quality guideline (SQG) has been developed for separation-center based system using anion column Metrosep A- singular ecosystems with freshwater or marine sediments. In ma- SUPP-5-150®). Precision was 10% for all dissolved nutrient analysis. rine ecosystems, the mean Effect Range Median quotients (m-ERM- q) is a pollutant-specific index from the SQG obtained in laboratory (Table 2) increased two fold and could be related to intense aqua- experiments using amphipod organisms (Long et al., 1998). It takes culture activity (Table 1a). During the rainy season, different dis- into account mixtures of contaminants, including trace metals, tributions are observed with lower nutrient concentrations present in sediment samples, and provides a management tool for highlighting the strong dilution capacity of the Mekong River. assessing sediment quality in terms of adverse biological effects. Despite significant progress in the development of the wastewater The m-ERM-q calculation is based on: management sector in Vietnam, 60% of urban wastewater is con- nected to sewerage and only 10% of this collected wastewater is P n ðC =ERM Þ treated (World Bank, 2013). Specifically in the Mekong Delta, water m —ERM —q ¼ i¼1 i i n quality management is an urgent problem with regards the mul- tiple sources of pollution including domestic, urban, agricultural, where Ciis the concentration of the pollutant i in the sample, ERMi industrial and aquaculture sources (Waibel et al., 2012). A dramatic fi is the experimentally de ned effect concentration for the pollutant rise in nutrient concentrations has also been observed on a global i and n is the number of studied pollutants i. Four classes of toxicity scale in the last decade, in response to land use changes and urban < probability for biota are then defined: low (m-ERM-q 0.1), low- development, in rivers from South-East Asia countries such as in e e medium (m-ERM-q: 0.11 0.5), medium high (m-ERM-q: the lower Mekong basin in Laos (Li and Bush, 2015), the Red River e > 0.51 1.5) and high priority sites (m-ERM-q 1.5) (Long et al., 1998). watershed in Northern Vietnam (Garnier et al., 2015), and in the The m-ERM-q index is derived from the concept of effect range-low Yangtze River in China (Li et al., 2011). (ERL i.e.10th percentile of the effect dataset) and effect range- In the Tien River, SPM concentrations are low during both sea- th fi median (ERM i.e. 50 percentile of the effect dataset) de ned by sons (Table 3) and in a similar range as previously reported by Noh Long et al. (1995) respectively as the concentrations below which et al. (2013). Higher concentrations were measured during the adverse effects are not expected to occur, above which adverse rainy seasons and are related to enhanced SPM transfer in the effects are expected to occur. Mekong watershed (Loisel et al., 2014). During the dry season, the In freshwater ecosystems, MacDonald et al. (2000) evaluated a low SPM concentrations increased sharply in the brackish zone consensus-based SQG for 28 chemicals (i.e. metals, polycyclic aro- (Table 3) suggesting estuarine flocculation processes (Lefebvre et matic hydrocarbons, polychlorinated biphenyls, and pesticides) and al., 2012). The POCSPM contents vary from 1.1% to 9.0% (Table 3) developed the threshold effect concentration (TEC) and probable and are in the same range as previously observed by Noh et al. effect concentration (PEC) indexes for freshwater sediment. The (2013). During the dry season, the higher POCSPM contents fi TEC was de ned as the concentration below which adverse effects measured in the riverine part support our hypothesis regarding the are not expected to occur and the PEC as the concentration above importance of local urban wastewater release into the river. Then, which adverse effects are expected to occur more often than not. in the brackish zone, POCSPM decreased (i.e. MD6 to MD9; Table 3) Based on the mean Effect Range Median quotient (m-ERM-q; Long suggesting both enhanced organic degradation processes in the fi et al., 1998), a mean PEC quotient (m-PEC-q) was de ned by estuarine water column and dilution of fluvial particles by organic- MacDonald et al. (2000) as predicting the absence of toxicity if depleted resuspended sediment (Etcheber et al., 2007; Statham, mean PEC quotient is lower than 0.5 and a high probability of 2012). During the rainy season, the low POCSPM content (mean of sediment toxicity if the mean PEC quotient is higher than 0.5. 1.3%) associated with high SPM concentration levels (Table 3) is likely related to the terrigenous origin of the SPM. This minimum 2.5. Statistical analysis POCSPM value, defined as the terrigenous signature, can range from 0.5% in a very erodible watershed (e.g. in the Taiwan watershed; The Pearson correlation coefficient matrix and one-way ANOVA Kao and Liu, 1997) and up to 5% in an equatorial forested watershed at probability level p < 0.05 (to test seasonal effect) were performed (e.g. the Congo River Coynel et al. (2005). using statistical package software (SPSS; version 23). 3.2. Filtered and suspended particulate metal dynamics in the Tien 3. Results & discussion River

3.1. River water mass characteristics Filtered (VF, CrF, CoF, NiF, CuF, ZnF, AsF, MoF, CdF, PbF, UF) and particulate (VSPM, CrSPM, CoSPM, NiSPM, CuSPM, ZnSPM, AsSPM, MoSPM, Physico-chemical parameters vary along the river and between CdSPM, PbSPM, AlSPM, FeSPM, MnSPM) trace metal concentrations the seasons (Table 1b). The dry season is characterized by higher measured in the Tien River (Table 2; Table 3) are within the mean water temperature (approximately þ 2 ○C) and a saline water world rivers dissolved (Gaillardet et al., 2014) and particulate (Viers intrusion up to the MD7 site (the intrusion being downstream et al., 2009) values and in the same concentration range as previ- during rainy season). The pH and DO values vary with opposite ously reported in the Mekong Delta (Cenci and Martin, 2004) and in seasonal trends: pH being more acidic during the rainy season the Lower Mekong River (Gaillardet et al., 1999). In particular, AsF whereas water is less oxygenated during the dry season. The Tien concentrations remain far below the concentrations previously River acidiﬁcation is closely related to the leaching of the sur- reported in the Mekong Delta aquifers (Berg et al., 2007). The rounding acid sulfate soils during rainy seasons (Minh et al., 1997) comparison to main Asian tropical rivers (Table 4) such as the while the decreasing oxygenation during the dry season could be Yangtze River Estuary (Yang et al., 2014), the Huanghe River Estuary the result of enhanced organic matter degradation along the river (Wang et al., 2016a) and the Pearl River Estuary (Zhang and Liu, network (Trinh et al., 2012). The local DO decrease observed at MD6 2002; Zhang et al., 2013) demonstrates that the Mekong Delta is during both seasons is the likely result of direct domestic and urban in the low range of both dissolved and particulate concentrations, discharges from My Tho City (230 000 inhabitants; Table 1a). conﬁrming the previous statement of Cenci and Martin (2004) on During the dry season, the P-PO4 concentrations are quite low but the negligible anthropogenic metal inputs to this system. The pH, increases locally in nearby urban centres (between MD-2 and MD- DO and SPM concentrations have a limited effect on the dynamic of 4; Table 1a, 2) while the N-NO3 concentrations are likely the result MetalF concentrations (r ¼ 0.73, p < 0.01 level for NiF and SPM only) of intense agricultural activity along this river (Table 2). In the and MetalSPM concentrations (r ¼ 0.79 and r ¼ 0.76, p < 0.01 level estuarine environment, (MD-8 to MD-10) N-NH4 concentrations for AlSPM and FeSPM with SPM only). Accordingly, the metal Table 1b Measured dissolved oxygen (DO), pH, temperature, conductivity, salinity and SPM concentrations in surface water, grain size distribution and modes of surface sediments at each sampling site during both dry and rainy seasons. (n.d.: not determined).

Sampling site D.O. mg L—1 pH T ○C Conductivity mS cm—1 Sal Grain size distribution

D10 D50 D90 mode

March dry season MD-1 4.82 7.82 31.8 0 .2 1 6 0.1 2.4 12.6 48.5 uni-modal MD-2 4.9 7.61 0 .2 0 7 0.1 5.6 49.3 1 1 7 .0 uni-modal MD-3 5.02 7.05 30.1 0.21 0.1 4.0 47.2 1 7 1 .5 bi-modal MD-4 5.06 7.44 30.2 0 .2 1 9 0.1 2.9 39.8 1 6 1 .4 bi-modal MD-5 4.67 7.59 0 .2 1 2 0.1 4.0 42.3 3 1 0 .8 bi-modal MD-6 3.77 7.58 30.3 0 .4 0 4 0.1 2.5 15.3 56.7 uni-modal MD-7 4.48 7.68 30.5 3.4 1.7 2.5 13.8 56.5 uni-modal MD-8 4.31 7.8 30.6 8.33 4.7 2.6 15.9 1 4 3 .8 bi-modal MD-9 6 7.91 n.d. 18.9 11.3 1.7 6.4 24.4 uni-modal MD-10 6 8.25 30.2 30.7 19 2.2 16.3 97.9 bi-modal October rainy season MD-4 4.88 7.22 27.8 80 0 2.3 11.9 46.2 uni-modal MD-5 5.31 7.1 27.7 75.5 0 1.7 7.6 31.9 uni-modal MD-6 4.63 6.95 27.9 82.5 0 2.7 16.4 57.8 uni-modal MD-7 5.41 7.1 27.8 84.7 0 1.5 6.5 21.2 uni-modal MD-10 6.6 7.4 27.9 3470 1.8 2.1 10.7 58.0 uni-modal

Table 2 Measured ﬁltered metal (VF, CrF, CoF, NiF, CuF, ZnF, AsF, MoF, CdF, PbF, UF) and nutrient (P-PO4, NeNH4) concentrations in surface water at each sampling site during both dry and rainy seasons. (n.d.: not determined).

Sites VF CrF CoF NiF CuF ZnF AsF MoF CdF PbF UF P-PO4 NeNH4 NeNO3

units mg L—1 mg L—1 mgL—1 mg L—1 mg L—1 mgL—1 mg L—1 mg L—1 mg L—1 mgL—1 mg L—1 mg L—1 mg L—1 mg L—1

March dry season MD-1 1.37 0 .0 9 2 0 .0 6 1 0.46 0.63 11.8 1.63 0.35 0 .0 0 5 0.041 0.38 0.02 0.06 1.5 MD-2 1.32 0 .0 6 6 0 .0 4 2 0.62 n.d. 9.3 1.60 0.36 0 .0 1 0 0.061 0.38 0.03 0.07 1.7 MD-3 1.28 0 .0 4 4 0 .0 4 0 0.46 0.78 14.6 1.50 0.36 0 .0 0 9 0.017 0.36 0.22 0.03 1.8 MD-4 1.16 0 .0 6 2 0 .0 4 7 0.49 1.06 30.2 1.30 0.32 0 .0 2 2 0.246 0.27 0.04 0.06 2.0 MD-5 1.30 0 .0 0 7 0 .0 3 1 0.44 0.64 11.7 1.38 0.32 0 .0 0 4 0.013 0.18 0.02 0.06 2.2 MD-6 2.31 0 .1 0 9 0 .0 6 4 0.48 0.99 72.3 1.38 0.27 0 .0 0 7 0.141 0.11 0.01 0.05 2.8 MD-7 n.d. 0 .9 7 8 0 .0 6 7 0.70 0.99 7.1 n.d. 0.67 0 .0 0 8 0.061 0.16 0.01 0.07 1.5 MD-8 1.43 0 .1 5 7 0 .0 4 6 0.52 1.22 18.3 1.16 1.63 0 .0 1 2 0.080 0.51 0.01 0.14 1.4 MD-9 1.67 0 .1 0 9 0 .0 4 1 0.61 1.21 12.9 1.36 3.56 0 .0 1 3 0.039 1.19 0.01 0.14 n.d. MD-10 1.95 0 .1 3 8 0 .0 3 8 0.73 1.41 32.2 1.47 6.37 0 .0 2 4 0.035 2.04 0.01 0.16 n.d. October rainy season MD-4 1.41 0 .0 5 9 0 .4 7 6 3.26 2.90 29.9 0.93 0.12 <0 .0 1 0.804 0.05 0.01 0.07 0.2 MD-5 1.38 0 .0 3 5 0 .5 2 4 2.22 0.96 37.4 0.88 0.52 <0 .0 1 0.116 0.04 0.01 0.03 0.2 MD-6 1.34 0 .3 1 3 1 .2 9 8 3.10 1.09 9.9 0.84 0.13 <0 .0 1 0.017 0.04 0.02 0.03 0.2 MD-7 1.28 0 .0 7 1 0 .4 4 2 2.30 0.96 20.5 0.77 0.13 <0 .0 1 0.063 0.05 0.01 0.02 0.2 MD-10 1.47 <0 .0 1 1 .3 3 2 3.73 3.56 4.9 2.57 0.39 <0 .0 1 <0.01 0.09 0.01 0.02 0.2

Table 3 Measured particulate metal (VSPM, CrSPM, CoSPM, NiSPM, CuSPM, ZnSPM, AsSPM, MoSPM, CdSPM, PbSPM, AlSPM, FeSPM, MnSPM) and particulate organic carbon (POC) concentrations in surface suspended sediment at each sampling site during both dry and rainy seasons. (n.d.: not determined).

Sites VSPM CrSPM CoSPM NiSPM CuSPM ZnSPM AsSPM MoSPM CdSPM PbSPM AlSPM FeSPM MnSPM POCSPM SPM

units Mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 mg kg—1 % mg L—1 March dry season MD-1 n.d. 18.0 6.8 12.1 13.9 n.d. n.d. n.d. n.d. 16.0 12 760 15 105 675 9.0 7 MD-2 n.d. 31.0 15.1 23.3 24.4 168 n.d. n.d. n.d. 27.8 21 827 27 909 1132 6.2 9 MD-3 n.d. 27.8 13.3 20.2 21.1 n.d. n.d. n.d. n.d. 24.0 20 281 25 035 1005 5.4 8 MD-4 n.d. 36.2 15.7 24.7 92.1 n.d. n.d. n.d. n.d. 29.0 23 538 30 528 753 3.5 11 MD-5 n.d. 38.0 17.7 28.0 27.1 335 n.d. n.d. n.d. 36.6 29 967 35 531 678 5.5 11 MD-6 n.d. 43.3 20.4 32.4 25.9 575 n.d. n.d. n.d. 27.9 34 044 35 902 569 1.7 43 MD-7 n.d. 41.1 23.1 34.5 27.5 229 n.d. n.d. n.d. 30.9 29 814 36 704 1147 1.4 133 MD-8 n.d. 39.7 18.5 29.2 33.6 92 n.d. n.d. n.d. 23.5 25 199 31 043 658 1.4 63 MD-9 n.d. 61.6 28.2 43.4 31.0 n.d. n.d. n.d. n.d. 39.0 39 778 47 598 1148 2.2 28 MD-10 n.d. 41.6 18.4 29.8 29.0 n.d. n.d. n.d. n.d. 25.2 26 304 31 389 721 2.7 30 October rainy season MD-4 125 23.4 14.1 27.3 41.0 147 21.6 0.45 0.17 32.7 99 599 52 591 297 1.1 154 MD-5 122 38.9 14.3 33.0 40.6 144 22.1 0.58 0.17 33.4 94 222 53 578 490 1.2 171 MD-6 136 34.8 16.4 34.4 47.1 287 24.3 0.61 0.28 37.0 106 488 58 989 552 1.3 148 MD-7 121 n.d. 13.4 16.1 36.9 139 21.4 0.27 0.23 32.9 104 556 54 938 85 1.4 128 MD-10 134 21.7 14.3 26.2 43.0 151 25.2 0.80

partitioning in the river water column was firstly assessed by the assessed in a different way by taking into account the total con- —1 partitioning coefficient Kd defined as the ratio of MetalSPM over centration relative to the volume of water, expressed in mg L (i.e. MetalF (Turner et al., 1993). The LogKd (Table 5) are stable in the the particulate concentrations were multiplied to the SPM con- Tien River and do not present any significant variation with pH, DO centrations and summed to the dissolved concentrations). The total and SPM (p > 0.05). Consequently, the metal partitioning was concentrations expressed in mgL—1 (Fig. 2) vary through the Tien Table 4 Dissolved and particulate (in both suspended and deposited sediment) trace metal concentrations of main Asian tropical rivers. (References cited:Cenci and Martin, 2004; Yang et al., 2 0 1 4 ; Wang et al., 2 0 1 6b; Zhang et al., 2 0 1 3 ; Zhang and Liu, 2 0 0 2 ; Wang et al., 2015, 2 0 1 7; Datta and Subramanian, 1998).

Location V Cr Co Ni Cu Zn As Mo Cd Pb Hg References

Dissolved concentrations ppb, min-max Mekong Delta 1.2e2.3 0.01 0.03 0.44 0.6 4 .9 e 7 2 0.8e2.6 0 .1 2 e6.4 0 .0 1 0 0.01 Current study e 0 .3 1 e 0 .0 7 e 0 .7 3 e 3 .6 e 0 .0 2 4 e 0 .8 0 Mekong Delta 0.13 0.2 0 .0 0 1 0.01 Cenci and Martin, 2004 e 0 .8 8 e 1 .5 e 0 .0 5 0 e 0 .1 6 Yangtze River 0 .0 3 e 23 0.01 0.06 0 .1 e 4 4 0 .3 e 130 0.1 0 .0 0 1 e1.9 7 0.01 Yang et al., 2014 Estuary e 2 3 .3 e 2 5 .9 e 243 e 3 .3 4 Huanghe River 0 .9 e 6 1 .2 e 35 1.4e2.5 0 .0 0 3 e0.3 2 0 .0 4 e8.2 Wang et al., 2016a Estuary Pearl River 0.3 3 .7 e 36 0.2e8.2 0 .0 0 1e0.3 0.19 Zhang et al., 2013 Estuary e 3 .3 e 4 .5 8 Suspended particulate concentrations ppm, min-max Mekong Delta 121 18e 62 6 .8 e 28 12e 43 14e 92 21e 25 0.27 0 .1 7 e0.28 16e 39 Current study e 136 e 0 .8 0 Mekong Delta 9 e 158 1 e 41 7 e 123 2 e 73 4 e 84 Cenci and Martin, 2004 Yangtze River 45e1273 4 .1 e71 12e204 22 54 3.9 0 .1 1 e54.9 6 e1445 Yang et al., 2014 Estuary e 973 e 3510 e 394 Huanghe River 0 .6 e 41 1 .5 e 88 0.1e7.9 0 .0 0 1 e0.2 3 2 e 79 0 .0 0 1 Wang et al., 2016a Estuary e 0 .2 2 4 Pearl River 71 18 62 51 212 0.82 75 Zhang and Liu 2002mean Estuary only Surface sediment particulate concentrations ppm, min-max Mekong Delta 39e 161 24e 87 7 .4 e 22 13e 48 7 .1 e 36 8 .4 e 21 39e 161 0 .0 8 e0.31 12e 31 0 .0 2 3 Current study e 0 .1 0 6 Yangtze River 50e 123 20e 42 9 .7 e 49 46e 127 5 .0 e 17 0 .0 7 e0.71 15e 45 Wang et al., 2015 Estuary Huanghe River 14e 64 1 .2 e 47 30e 81 4 .1 e 23 0 .0 4 e0.2 3 .2 e 34 0 .0 1 2 Wang et al., 2017 Estuary e 0 .3 7 9 Pearl River 10e 89 52e 186 10e 23 0 .0 4 e0.84 20e 64 Zhang et al., 2013 Estuary Ganges 36e 50 55e 146 18e 39 15e 31 53e 117 14e 24 Datta and Subramanian, River 1998 Brahmaputra 43e 86 55e 160 18e 51 15e 51 48e 149 10e 38 Datta and Subramanian, River 1998

River and are higher during the wet season than during the dry been studied, UF behaves conservatively (Windom et al., 2 0 00; season (ANOVA, p < 0.01 for Cr, Cu and Pb) suggesting enhanced reference within; Strady et al., 2009) while such behaviour is less metal load transport during the rainy season towards the ocean. usual for MoF and VF (e.g. in the Chao Phraya Estuary; Dalai et al., —1 Furthermore, particulate concentrations expressed in mg L vary 2005). Both MoF and VF can be affected by intra-estuarine trans- from site to site and are closely related to SPM concentration var- formation processes like SPM or sediment sorption/desorption re- iations (Fig. 2; p < 0.01 for total V, Cr, Co, Ni, Cu and Pb during both actions and biological uptake as observed in the Gironde Estuary seasons). This pattern supports that SPM concentrations control the (Strady et al., 2009) and in the Taiwan Strait (Wang et al., 2016a). load of total metal concentrations in the Mekong Delta during both Thus, this non-conservative behaviour deriving from the theoret- dry and wet seasons, as observed in the Yangtze River Estuary (Yang ical dilution line (i.e. gain or loss of dissolved element), illustrates a et al., 2014). Given the perspective of increased hydropower dam higher reactivity between the dissolved and particulate phase (e.g. constructions on the main channel and its tributaries, the resulting SPM or sediment). In the Tien River, gains of NiF, CuF, AsF, CoF and decrease of sediment load will probably affect the trace metal load CdF were observed (Fig. 3) suggesting intense desorption processes in the Mekong Delta. During transportation, the particulate phase is from particles occurring in the salinity gradient. This bell-shaped the dominant phase for V, Cr, Co, As and Pb (Fig. 2) while the dis- pattern is well documented for CdF as a chloride-induced desorp- solved metal fraction dominates for Mo, Ni and Cu in the riverine tion from the suspended sediments and has been previously part, and only for Mo in the estuarine part (Fig. 2). Dissolved metal observed in the Mekong Delta (Cenci and Martin, 2004) and other concentrations vary along the estuarine salinity gradient (Table 2) estuaries like the Gironde Estuary (Kraepiel et al., 1997; Dabrin et resulting in high particulate-dissolved reactivity associated with al., 2009), the Pearl River (Wang et al., 2012a), the Saigon River physical, chemical and biological processes in the mixing zone (de (Strady et al., 2017) and the Huanghe River Estuary (Wang et al., Souza Machado et al., 2016). This reactivity can be approached by 2016b). For CuF, even if a conservative behaviour is more usual and the mixing diagram (e.g. Liss, 1976; de Souza Machado et al., 2016), has been observed in the Mekong Delta (e.g. Cenci and Martin, which reflects the concentration of an element over salinity be- 2004), in the Pearl River (Wang et al., 2012a) and in the Huanghe tween two endmembers: the river and the ocean (i.e. data from the River estuary (Wang et al., 2016b), desorption from the reactive South China Sea (Censi and Martin, 2004; Wang et al., 2012b; Wen particles is the suggested explanation for the bell-shaped pattern et al., 2006) and from mean ocean values (Nozaki, 2001)). When the observed in the Gironde Estuary (Kraepiel et al., 1997). In the Tien concentrations follow the theoretical dilution line between the two River, this hypothesis could explain the gain of CuF but the absence endmembers, the element is thus controlled by the physical mixing of significant correlation with SPM concentrations along the system and its behaviour is defined as conservative. Such conservative (p > 0.05) and between LogKd and salinity (p > 0.05) does not allow behaviour is observed for UF, VF, and MoF in the Tien River (Fig. 3; us to make any conclusions on Cu particle reactivity. Unlike pre- p < 0.01 for UF and MoF with salinity). In most estuaries that have vious measurements performed in the Mekong Delta (Cenci and Fig. 2. Spatial variation of calculated total metal concentrations (sum of dissolved and particulate concentrations; V, Cr, Co, Ni, Cu, As, Mo, Pb) expressed in mg L—1 and SPM concentrations expressed in mg L—1 during both dry and rainy seasons.

Martin, 2004) and the Huanghe River estuary (Wang et al., 2016b), during organic degradation processes occurring in the estuary AsF and NiF concentrations increase in the mid salinity range. In the (Table 2; 3). For NiF, the signiﬁcant negative correlation with SPM case of As, the linear positive correlation (r ¼ 0.79) observed be- concentrations (r ¼ 0.74, p < 0.01) suggests desorption from sus- tween POCSPM and AsF during the dry season suggests an Asrelease pended particles along the salinity gradient. This is in accordance

Fig. 3. Filtered metal concentrations along the salinity gradient a) V, Mo, U; b) Ni, Cu, As; c) Cd, Co; d) Cr, Pb and Zn (expressed as Zn/100) during both seasons. The marine endmember is characterized according to data from Wang et al., 2012b; Wen et al., 2006; Censi and Martin, 2004, Nozaki, 2001. with reported laboratory experiments and modelling of metal (r ¼ 0.76 and 0.76 p < 0.01 Table 7) suggest that organic matter is a desorption kinetics which highlight Ni desorption from polluted major carrier for these elements, as it has been observed for Cu in riverine and estuarine sediments (Millward and Liu, 2003). How- the Saigon River sediments (Strady et al., 2017) and for Hg in the ever, the impact of desorption on observed non-conservative Lower Mekong Basin sediments (Noh et al., 2013; Gue´dron et al., behaviour was minimized by the authors because of the low Ni 2014). In the particular case of As, the positive correlation kinetic desorption and the non-attempt of sorption equilibrium in a observed between AsSED and POCSED is unusual since As has been dynamic estuarine environment. Finally, concerning CrF, ZnF and documented to be primarily controlled by sorption onto metal PbF, the high variability of measured concentrations in the riverine oxide surfaces in natural environments (e.g. Redman et al., 2002) part make it difficult to get a proper river endmember and therefore although it has been reported that As was sequestered as organo- to determine the gain or loss of metal towards the ocean. Further arsenic complexes in organic rich sediments (Baruah et al., 2003). investigation is needed to find the origin of this high variability, However, the co-precipitation of microbially synthetized arseno- either from anthropogenic inputs (e.g. small localised industrial pyrites on organic matter surfaces or on the formation of organo- activity such as electronics, textiles and construction) or from acid arsenic complexes (Paikaray et al., 2005) can both explain the un- sulfate soil leaching. usual observed correlation. Finally, MetalSED are positively correlated to FeSED and MnSED, 3.3. Factors controlling metal distributions in the Tien River surface with a significant correlation (p < 0.01) between FeSED and the sediments following MetalSED in the specific sequence: PbSED > NiSED > AsSED > CoSED > CuSED > AlSED > HgSED The surface sediments of the Tien River present a grain size > MnSED > CrSED > ZnSED > VSED and between MnSED and the distribution dominated by silt classes (54e87%) and a median following MetalSED in the specific sequence: FeSED > PbSED > CoSED grain-size (D50) comprised between 6 and 49 mm (Table 1b). The > HgSED > CuSED > AlSED > AsSED > ZnSED > NiSED. > CrSED > VSED. The grain size distributions varied from site to site and with seasons, sorption selectivity on most metal-oxide minerals determined by without exhibiting a specific spatial pattern from the river to the Schultz et al. (1987) follows the sequence coastline. The POCSED content ranges from 0.2 to 1.5% (Table 6). Cr ≤ Pb ≤ Cu > Co ≤ Zn > Ni ≤ Cd, with some distinct differences MetalSED concentrations measured in the Tien River (Table 6) are in such as the preference of MnO2 for Co (Smith, 1999). The sequence the same range as previously observed in this system (Cenci and based on MetalSED and FeSED/MnSED correlation in the Tien River Martin, 2004; Noh et al., 2013) and are in the low concentration surface sediments differs and exhibits a specific pattern, which range of the main Asian Rivers (Table 4) such as the Yangtze River could be related to the presence and characteristics of the sur- Estuary (Wang et al., 2015), the Huanghe River Estuary (Wang et al., rounding acid sulfate soils (Minh et al., 1997). The oxidation of these 2017), the Pearl River Estuary (Zhang et al., 2013), the Ganges- soils causes metals to be redistributed from the “pyritic” and Brahmaputra Rivers (Datta and Subramanian, 1998) and including “organic” fractions to the “acid-soluble” fraction (Claff et al., 2011). the Saigon River (Strady et al., 2017). Cenci and Martin (2004) Then, the subsequent acidification, due to exceedance of the acid observed that MetalSED in the Mekong Delta varies spatially but neutralizing capacity of the soil, drives the release of metals to the not seasonally. In this study, we observed the same pattern except “labile” fraction (Claff et al., 2011). Thus, the oxidation and acidifi- for CuSED and ZnSED (ANOVA, p > 0.05). cation of sulfidic soil could lead to changes in metal mobility and In surface sediments, the Pearson correlation matrix applied to consequently to metal sorption phases and distribution in the both seasons highlights positive correlations between the entire sediments. In the Tien River, the comparison of MetalSED with MetalSED set (including AlSED, MnSED and FeSED), and also between MetalSPM for both seasons gives similar concentration ranges MetalSED and POCSED (Table 7). These results support that (i) (Tables 3 and 6), except for Cr and Pb which present higher con- particle-size, (ii) POC content and/or (iii) Fe, Al and Mn e oxy(hydr) centrations in sediments and SPM respectively. Such differences for oxides are the major carrier phases controlling the spatial distri- Cr and Pb are likely related to a change in speciation and carrier bution of particulate trace metals in this system. Firstly, the small phases in the sediment resulting in diagenetic processes (e.g. a drop sized particles’ highly specific surface plays a major role in metal in pH and Eh). Because Cr is sensitive to redox states and oxygen- retention (e.g. Fo€rstner and Wittmann, 1979; Loring, 1991; Jung et ation levels, the difference between measured CrSED and CrSPM al., 2014). Secondly, although POCSED content is low (0.2%e could originate from differences between redox and oxygen values 1.5%; Table 6), its strong correlation with HgSED, AsSED (r ¼ 0.83 and in the two compartments, even though they were not measured at 0.84, for Hg and As respectively p < 0.01 Table 7) and CoSED, CuSED the water-sediment interface. Unlike in the sediment no significant

Table 5 LogKd (Cr, Co, Ni, Cu, As, Mo, Pb) at each sampling site during both dry and rainy seasons. (n.d.: not determined).

Sites LogKDCr LogKDCo LogKDNi LogKDCu LogKDAs LogKDMo LogKDPb

March dry season MD-1 5.3 5.0 4.4 4.3 n.d. n.d. 5.6 MD-2 5.7 5.6 4.6 n.d. n.d. n.d. 5.7 MD-3 5.8 5.5 4.6 4.4 n.d. n.d. 6.1 MD-4 5.8 5.5 4.7 4.9 n.d. n.d. 5.1 MD-5 n.d. 5.8 4.8 4.6 n.d. n.d. 6.4 MD-6 5.6 5.5 4.8 4.4 n.d. n.d. 5.3 MD-7 n.d. 5.5 4.7 4.4 n.d. n.d. 5.7 MD-8 5.4 5.6 4.7 4.4 n.d. n.d. 5.5 MD-9 5.8 5.8 4.9 4.4 n.d. n.d. 6.0 MD-10 5.5 5.7 4.6 4.3 n.d. n.d. 5.9 October rainy season MD-4 5.6 n.d. n.d. 4.2 4.4 3.6 4.6 MD-5 6.1 n.d. n.d. 4.6 4.4 3.0 5.5 MD-6 5.0 n.d. n.d. 4.6 4.5 3.7 6.3 MD-7 n.d. n.d. n.d. 4.6 4.4 3.3 5.7 MD-10 n.d. n.d. n.d. 4.1 4.0 3.3 n.d. Table 6 Measured particulate metal (VSED, CrSED, CoSED, NiSED, CuSED, ZnSED, AsSED, MoSED, CdSED, PbSED, AlSED, FeSED, MnSED) and particulate organic carbon (POC) concentrations in surface sediment at each sampling site during both dry and rainy seasons. (n.d.: not determined).

Sites VSED CrSED CoSED NiSED CuSED ZnSED AsSED MoSED CdSED PbSED HgSED AlSED FeSED MnSED POCSED

units mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 mg kg-1 % March dry season MD-1 110 71.4 16.7 37.0 36.2 171 19.0 0.57 0.31 30.4 0.087 79 626 46 884 1237 1.2 MD-2 65 48.0 11.0 22.6 20.7 110 11.3 0.46 0.17 16.8 0.047 44 947 28 627 642 0.6 MD-3 70 47.2 11.1 24.4 20.7 109 12.2 0.28 0.20 18.1 0.053 56 738 31 103 714 0.8 MD-4 39 24.5 7.4 13.4 7.1 64 8.4 0.07 0.08 12.2 0.023 32 926 19 808 287 0.2 MD-5 69 79.1 15.7 33.1 18.6 114 12.4 4.88 0.19 19.9 0.046 54 501 34 035 942 0.7 MD-6 81 78.0 17.0 37.4 33.5 169 17.3 0.66 0.22 29.5 0.106 57 177 45 837 1132 1.2 MD-7 98 76.6 16.6 39.7 28.9 155 18.8 0.88 0.14 25.7 0.074 70 907 47 073 1230 1.2 MD-8 88 73.9 14.9 32.9 21.8 135 13.6 0.47 0.13 22.9 0.056 50 708 38 666 825 0.7 MD-9 112 87.5 18.1 43.7 30.6 169 18.1 1.38 0.11 27.0 0.075 93 633 49 924 1172 1.1 MD-10 68 48.8 15.3 27.5 17.7 117 14.9 0.52 0.11 17.7 0.066 52 346 36 196 827 1.4 October rainy season MD-4 161 83.8 21.8 37.7 34.2 256 19.1 0.62 0.24 29.6 0.089 79 857 44 667 1216 1.2 MD-5 99 62.5 17.9 39.1 36.2 185 20.5 0.93 0.24 30.6 0.091 75 141 46 466 1328 1.5 MD-6 99 66.7 16.3 35.2 29.1 169 18.7 0.66 0.22 27.3 0.071 61 951 41 198 1076 0.8 MD-7 120 86.0 17.5 47.7 35.5 184 20.8 0.94 0.27 24.3 0.079 n.d. 41 977 805 1.3 MD-10 102 79.0 17.7 39.5 27.8 179 16.2 0.87 0.09 28.1 0.057 71 390 45 932 866 1.0

Table 7 —1 Pearson matrix correlation of particulate metal (VSED, CrSED, CoSED, NiSED, CuSED, ZnSED, AsSED, MoSED, CdSED, PbSED, AlSED, FeSED, MnSED; in mg kg ) and POCSED content (%) in sediment during both seasons.

VSED CrSED CoSED NiSED CuSED ZnSED AsSED MoSED CdSED PbSED HgSED AlSED FeSED MnSED POCSED

VSED 1 ** CrSED .7 5 5 1 ** ** CoSED .8 7 4 .8 6 5 1 ** ** ** NiSED .7 7 0 .9 1 9 .873 1 ** ** ** ** CuSED .8 1 1 .7 3 7 .821 .8 6 2 1 ** ** ** ** ** ZnSED .9 5 6 .7 4 9 .922 .7 8 4 .8 7 5 1 ** ** ** ** ** ** AsSED .8 0 7 .7 0 6 .851 .8 9 0 .9 3 3 .8 5 0 1

MoSED —.056 .382 .202 .209 —.065 —.063 —.060 1 ** * * CdSED .495 .330 .405 .402 .6 8 9 .5 1 7 .5 6 7 .059 1 ** ** ** ** ** ** ** * PbSED .7 7 6 .7 5 3 .861 .8 2 2 .9 2 5 .8 7 4 .8 7 1 —.010 .525 1 .6 6 8 ** .6 1 9 * .782 ** .7 3 1 ** .9 0 3 ** .7 9 0 ** .8 6 2 ** .620 * .8 5 4 ** 1 HgSED —.126 ** ** ** ** ** ** ** ** * AlSED .8 2 7 .7 3 9 .807 .8 6 6 .8 2 1 .7 8 5 .8 3 5 .074 .359 .8 0 8 .660 1 ** ** ** ** ** ** ** ** ** ** FeSED .7 5 4 .8 2 6 .888 .9 0 8 .8 8 2 .8 1 0 .8 9 0 .038 .346 .9 3 1 .830 .8 7 8 1 ** ** ** ** ** ** ** ** ** ** ** MnSED .6 8 2 .6 9 2 .837 .7 4 8 .8 3 1 .7 5 6 .8 2 9 .170 .505 .8 8 7 .836 .8 3 0 .8 9 3 1 * * ** ** ** ** ** ** ** ** ** ** POCSED .5 9 6 .5 3 1 .756 .7 2 9 .7 5 9 .6 6 2 .8 3 7 —.054 .406 .6 6 6 .828 .6 9 6 .7 8 6 .7 4 5 1 *Correlation is signiﬁcant at the 0.05 level (2-tailed). **Correlation is signiﬁcant at the 0.01 level (2-tailed).

correlation was observed between FeSPM and CrSPM although Fe moderate to strong contaminated sediment for As (2 < Igeo<4). concentrations are in the same range in both compartments. The Thus, these two indexes are showing a ZnSED, and to a greater extent change of affinity with Fe should be emphasized during field an AsSED enrichment or contamination issue in this environment. studies and/or experiments to point out the role of redox processes First of all, we would like to point out that the index conclusions on Cr partitioning. The observed lower PbSED concentrations when strongly depend on the background values chosen, i.e. the UCC compared to PbSPM could be the result of intense sediment lixivi- (Wedepohl, 1995) in the present study. In the case of the Mekong ation by estuarine and marine waters leading to increasing Pb River watershed, the geology is diverse and rich making it difficult mobility from the sediment to the dissolved phase as it has been to get a proper background value. In the absence this background observed in the Yangtze River Estuary (Zhao et al., 2013). This hy- value from a sediment core, Wang et al. (2012b) defined a back- pothesis cannot be supported in the Tien River. ground value from a geometric means of a high number of samples from the Manwan Reservoir in the Upper Mekong River in China. However, this calculated background value presents high coeffi- 3.4. Risk assessment of Tien River surface sediments cient of variation and low representativity, which can be an issue if applying to the Mekong Delta, considering the distance between To assess the degree of metal enrichment in the surface sedi- this reservoir and the Mekong Delta (up to 2,600 km) and the high ments of the Tien River we compared two geochemical indexes: the number of confluent rivers with specific geological characteristics Enrichment Factor (EF) and the Geoaccumulation Index (Igeo) (see of the catchment area. Thus, we consider that the UCC is the most material and method 2.3.), calculated for each site and metal. suitable background value for this environment. The conclusion of Globally, the calculated EF at all sites (Table 8) evidences none to As enrichment in the Tien River is coherent with the literature minor enrichment for V, Cr, Co, Ni, Cu, Mo, Cd, Pb and Hg (EF < 1.5), SED on As contamination in the Mekong Delta, characterized by high moderate enrichment for Zn (3 < EF < 5), and moderately severe to levels of As in groundwater and in sediment cores and also by a severe enrichment for As (5 < EF < 25). The calculated Igeo at all serious health risk to the local population (e.g. Berg et al., 2007; sites (Table 8) demonstrates uncontaminated sediment to minor Buschmann et al., 2008; Hoang et al., 2010). Note that because of contamination for V, Cr, Co, Ni, Cu, Mo, Cd, Pb and Hg (Igeo<1), the poor Zn accuracy measurement and the consequences on Zn moderate contaminated sediments for Zn (1 < Igeo<2), and Table 8 Calculated EF and Igeo for surface sediment at each sampling site during both dry and rainy seasons. (n.d.: not determined).

Enrichment Factor EFV EFCr EFCo EFNi EFcu EFZn EFAs EFMo EFcd EFPb EFHg

March dry season MD-1 2.0 2.0 1.4 1.9 2.5 3.2 9.2 0.4 3.0 1.7 1.5 MD-2 2.1 2.4 1.6 2.1 2.5 3.6 9.8 0.6 2.8 1.7 1.5 MD-3 1.8 1.8 1.3 1.8 2.0 2.9 8.3 0.3 2.7 1.5 1.3 MD-4 1.7 1.6 1.5 1.7 1.2 2.9 9.9 0.1 1.9 1.7 0.9 MD-5 1.8 3.2 1.9 2.5 1.9 3.1 8.8 n.d. 2.7 1.7 1.2 MD-6 2.1 3.0 2.0 2.7 3.2 4.4 11.7 0.6 2.9 2.3 2.6 MD-7 2.0 2.4 1.6 2.3 2.2 3.2 10.3 0.7 1.5 1.6 1.4 MD-8 2.5 3.2 2.0 2.7 2.3 4.0 10.4 0.5 2.0 2.1 1.5 MD-9 1.7 2.1 1.3 1.9 1.8 2.7 7.5 0.8 0.9 1.3 1.1 MD-10 1.9 2.1 1.9 2.2 1.8 3.3 11.0 0.5 1.5 1.5 1.7 3.0 2.3 1.8 2.0 2.3 4.8 9.2 2.3 1.7 1.5 October rainy season MD-4 0.4 1.9 1.8 1.6 2.2 2.6 3.7 10.6 2.5 1.9 1.7 MD-5 0.7 2.3 2.4 1.8 2.4 2.5 4.1 11.7 0.6 2.7 2.0 1.6 MD-6 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. MD-7 MD-10 2.1 2.4 1.7 2.3 2.1 3.7 8.8 0.7 1.0 1.8 1.1 Igeo Ig e o IgeoV IgeoCr IgeoCo IgeoNi Igeocu IgeoZn IgeoAs Mo Igeocd IgeoPb IgeoHg

March dry season MD-1 0.5 0.4 —0.1 0.4 0.8 1.1 2.7 —1.9 1.0 0.3 0.0 MD-2 —0.3 —0.1 —0.7 —0.3 —0.1 0.5 1.9 —2.2 0.1 —0.6 —0.8 MD-3 —0.2 —0.2 —0.6 —0.2 —0.1 0.5 2.0 —2.9 0.4 —0.5 —0.7 MD-4 —1.0 —1.1 —1.2 —1.1 —1.6 —0.3 1.5 —5.0 —0.9 —1.1 —1.9 MD-5 —0.2 0.6 —0.1 0.2 —0.2 0.5 2.1 n.d. 0.3 —0.4 —0.9 MD-6 0.0 0.6 0.0 0.4 0.6 1.1 2.5 —1.7 0.5 0.2 0.3 MD-7 0.3 0.5 —0.1 0.5 0.4 1.0 2.6 —1.3 —0.1 0.0 —0.2 MD-8 0.1 0.5 —0.2 0.2 0.0 0.8 2.2 —2.2 —0.2 —0.2 —0.6 MD-9 0.5 0.7 0.1 0.6 0.5 1.1 2.6 —0.6 —0.4 0.1 —0.2 2.3 MD-10 —0.2 —0.1 —0.2 0.0 —0.3 0.6 —2.0 —0.5 —0.5 —0.4 2.7 October rainy season MD-4 1.0 0.7 0.3 0.4 0.7 1.7 —1.8 0.7 0.2 0.1 2.8 MD-5 0.3 0.3 0.0 0.5 0.8 1.2 —1.2 0.7 0.3 0.1 2.6 MD-6 0.3 0.3 —0.1 0.3 0.4 1.1 —1.7 0.5 0.1 —0.2 0.8 1.2 2.8 MD-7 0.6 0.7 0.0 0.7 —1.2 0.8 —0.1 —0.1 0.5 1.2 2.4 MD-10 0.4 0.6 0.0 0.4 —1.3 —0.7 0.1 —0.5

concentration over-estimation, the observed ZnSED enrichment SQG values for freshwater sediment (i.e MD-1 to MD-7) using TEC, may be either over-estimated or may originate from the direct PEC and m-ERM-q indexes and for marine sediments (i.e. MD-7 to release of industrial and urban wastewater without prior treatment MD-10) using ERL, ERM and m-ERM-q indexes (Table 9). Both and from urban release such as street and construction dust, house freshwater and marine sediments exhibit CdSED, PbSED, HgSED con- roof made of zinc, or worn out tires. centration ranges below which adverse effects are not expected to According to the metal enrichment index results in the Tien occur (lower than TEC and ERL) while CrSED, NiSED, CuSED, ZnSED, River we assessed the ecotoxicological status of its sediments by AsSED exhibit a concentration range above which adverse effects are comparing the measured concentrations with the commonly used expected to occur (Table 9). The possible adverse effects of toxic mixtures (Cr, Ni, Cu, Zn, As, Cd, Pb, Hg) evaluated by the calculation of the m-PEC-q and m-ERM-q for freshwater and marine sediments Table 9 respectively (Table 9), predict an absence of toxicity in the surface Calculated sediment quality guidelines index for freshwater (TEC, PEC, m-PEC-q) freshwater sediments and possible low to medium ecotoxicological and marine water (ERL, ERM, m-ERM-q) ecosystems for surface sediment at each potential effects in the estuarine-marine sediments of the Tien sampling site during both dry and rainy seasons. N is the number of samples River. Because of possible Zn concentration over-estimation, m- comprised between TEC and PEC values and between ERL and ERM values for each metal. PEC-q and m-ERM-q were calculated excluding ZnSED. The two calculations present similar index values (data not shown) sug- Cr Ni Cu Zn As Cd Pb Hg gesting the low contribution of Zn to possible adverse toxic effects. TEC 43 22.7 32 121 9.8 1.0 35.8 0.18 These results support the idea that Mekong River sediments pre- PEC 111 48.6 149 459 33 5.0 128 1.06 sent a low risk to benthic organisms according to metal contami- N TEC < Tien River < PEC 11 9 5 9 10 0 0 0 ERL 81 20.9 34 150 8.2 1.2 46.7 0.15 nation. However, assumptions regarding ecotoxicological effects ERM 370 51.6 270 410 70 9.6 218 0.71 should be investigated in further studies combining metal accu- N ERL < Tien River < ERM 2 5 1 4 6 0 0 0 mulation in sediments and benthic organisms. MD- MD- MD- MD- MD- MD- MD- MD- MD- MD- 1 2 3 4 5 6 7 8 9 10 4. Conclusions

m-PEC-q freshwater sediments March 2013 dry 0.37 0.23 0.24 0.13 0.30 0.37 0.36 Despite the development of intense agricultural and aquacul- season ture activity, and of rapid urban growth illustrated by localised high October 2013 rainy 0.41 0.38 0.35 0.42 nutrient and organic carbon levels, the Tien River remains in a state season m-ERM-q estuarine and marine sediments of good quality regarding trace metal contamination. Trace metal March 2013 dry 0.25 0.20 0.26 0.18 distributions vary spatially and the metal partitioning appears to be season highly dynamic in the salinity gradient of the Mekong Delta. October 2013 rainy 0.29 0.25 Although the ecotoxicological indexes demonstrated a low to me- season dium contamination of the surface sediments, direct accumulation measurement in benthic organisms are required to assess the metal Pistocchi, C., Aissa-Grouz, N., Luu, T.N.M., Vilmin, L., Dorioz, J.-M., 2015. Phos- risk in biota. phorus budget in the water-agro-food system at nested scales in two contrasted regions of the world (ASEAN-8 and EU-27). Glob. Biogeochem. Cycles 29, 1348e13 68. Acknowledgments General Statistics Office of Vietnam GSO, 2016. Population and Employment, Sta- tistical Yearbook of Vietnam 2015. Statistical Publishing House, Hanoi, pp. 57e155. This study has been conducted under the framework of CARE- Giang, C.N.D., Sebesvari, Z., Renaud, F.G., Rosendahl, I., Minh, Q.H., Amelung, W., RESCIF initiative. The authors would like to thank the students for 2015. Occurrence and dissipation of the antibiotics sulfamethoxazole, sulfa- diazine, trimethoprim, and enrofloxacin in the Mekong Delta, Vietnam. PLoS their help during sampling and the anonymous reviewers for their One 10, 7. comments. This study was funded by incentive grants from the Gue´dron, S., Grangeon, S., Lanson, B., Grimaldi, M., 2 009 . Mercury speciation in a French National Research Institute for Sustainable Development tropical soil association; consequence of gold mining on Hg distribution in IRD, by grants from the Vietnam National University (grant number French Guiana. Geoderma 153, 331e346. Gue´dron,S., Tisserand, D., Garambois, S., Spadini, L., Molton, F., Bounvilay, B., 2014. B2016-20-05), and by grants from Re´gion Rho^ne Alpes COOPERA Baseline investigation of (methyl)mercury in waters, soils, sediments and key “Take Care”. foodstuffs in the lower Mekong Basin: the rapidly developing city of Vientiane (Lao PDR). J. Geochem Explor 143, 96e 102. Hoang, T.H., Bang, S., Kim, K.W., Nguyen, M.H., Dang, D.M., 2010. Arsenic in References groundwater and sediment in the Mekong River delta, Vietnam. Environ. Poll. 158, 2648e 2 65 8. Anonymous, 23 October, 2 00 0 . Directive of the European Parliament and of the Jung, H.S., Lim, D., Xu, Z., Kang, J.H., 2014. Quantitative compensation of grain-size Council (2000/60/EC) Establishing a Framework of Community Action in the effects in elemental concentration: a Korean coastal sediments case study. Field of Water Policy. Estuar. Coast. Shelf S 151, 69e77. Kao, S.J., Liu, K.K., 1997. Loads of dissolved and nonfossil particulate organic carbon APHA, 1995. Standard Methods for the Examination of Water and Wastewater. from an Oceania small river (Lanyang Hsi) in Taiwan. Biogeochem 39, 255e 2 69 . American Public Health Association, Washington, DC. Kraepiel, A.M.L., Chiffoleau, J.F., Martin, J.M., Morel, F.M.M., 1997. Geochemistry of Apitz, S.E., Power, E.A., 2 00 2. From risk assessment to sediment management: an trace metals in the Gironde Estuary. Geochim. Cosmochim. Acta 61, 1421e1436. international perspective. J. Soil Sed. 2, 61e66. Baruah, M.K., Kotoky, P., Baruah, J., Borah, G.C., Bora, P.K., 2 00 3 . Arsenic association Kummu, M., Lu, X.X., Wang, J.J., Varis, O., 2010. Basin-wide sediment trapping ef- and distribution in carbonaceous materials in northeastern India. Curr. Sci. ficiency of emerging reservoirs along the Mekong. Geomorph 119, 181e197. fi ´ India 85, 2 e 25. Lefebvre, J.P., Ouillon, S., Vinh, V.D., Ar , R., Panche , J.Y., Mari, X., Thuoc, C.V., ´ Berg, M., Stengel, C., Trang, P.T.K., Pham, H.V., Sampson, M.L., Leng, M., Samreth, S., Torre ton, J.P., 2012. Seasonal variability of cohesive sediment aggregation in the e Fredericks, D., 2007. Magnitude of arsenic pollution in the Mekong and Red Bach Dang-Cam Estuary, Haiphong (Vietnam). Geo-Mar. Lett. 32, 103 121. River Deltas- Cambodia and Vietnam. Sci. Total. Environ. 372, 413e425. Li, X., Yang, L., Yan, W., 2011. Model analysis of dissolved inorganic phosphorus Birth, G.A., 2 0 0 3 . A scheme for assessing human impacts on coastal aquatic envi- exports from the Yangtze river to the estuary. Nutr. Cycl. Agroecosyst 90, e ronments using sediments. In: Woodcoffe, C.D., Furness, R.A. (Eds.), Coastal GIS 157 170. Li, S., Bush, R., 2015. Rising flux of nutrients (C, N, P and Si) in the lower Mekong 2003. Wollongong University Papers in Centre for Maritime Policy, 14,Australia. River. J. Hydrol. 530,447e461. Buschmann, J., Berg, M., Stengel, C., Winkel, L., Sampson, M.L., Pham, T.K.T., Pham, H.V., 2 0 0 8 . Contamination of drinking water resources in the Mekong Liss, P.S., 1976. Conservative and non-conservative behavior of dissolved constitu- ents during estuarine mixing. In: Burton, J.D., Liss, P.S. (Eds.), Estuarine Chem- delta floodplains: arsenic and other trace metals pose serious health risk to e population. Environ. Intern 34, 756e 764. istry. Academic Press, London, pp. 93 1 30. Campbell, I.C., 2012. Biodiversity in the Mekong delta. In: Renaud, F.G., Künzer, C. Loisel, H., Mangin, A., Vantrepotte, V., Dessailly, D., Dinh, D.N., Garnesson, P., ´ (Eds.), The Mekong Delta System: Interdisciplinary Analyses of a River Delta. Ouillon, S., Lefebvre, J.-P., Me riaux, X., Phan, T.M., 2014. Variability of suspended ’ fl Springer Environ Sci Engineer, pp. 293e 3 14 (Chapter 11). particulate matter concentration in coastal waters under the Mekong s in u- Cenci, R.M., Martin, J.M., 2004. Concentration and fate of trace metals in Mekong ence from ocean color (MERIS) remote sensing over the last decade. Rem. Sens. River Delta. Sci. Total. Environ. 332, 167e182. Environ. 150, 218e230. Chapman, P.M., Wang, F., 2001. Assessing sediment contamination in estuaries. Long, E.R., Field, L.J., MacDonald, D.D., 1998. Predicting toxicity in marine sediments Environ. Toxicol. Chem. 20, 3 e 22. with numerical sediment quality guidelines. Environ. Toxicol. Chem. 17, Claff, S.R., Burton, E.D., Sullivan, L.A., Bush, R.T., 2011. Metal partitioning dynamics 714e727. during the oxidation and acidification of sulfidic soil. Chem. Geol. 286, 146e157. Long, E.R., MacDonald, D.D., Smith, S.L., Calder, F.D., 1995. Incidence of adverse Coynel, A., Seyler, P., Etcheber, H., Meybeck, M., Orange, D., 2 0 0 5 . Spatial and sea- biological effects within ranges of chemical concentrations in marine and sonal dynamics of total suspended sediment and organic carbon species in the estuarine sediments. Environ. Manag. 19, 81e97. Loring, D.H., 1991. Normalization of heavy-metal data from estuarine and coastal Congo River. Glob. Biogeochem. Cycles 19, GB4019. e Dabrin, A., Scha€fer, J., Blanc, G., Strady, E., Masson, M., Bossy, C., Castelle, S., sediments. ICES J. Mar. Sci. 48, 101 115. Girardot, N., Coynel, A., 200 9. Improving estuarine net flux estimates for dis- MacDonald, D.D., Ingersoll, C.G., Berger, T.A., 2000. Development and evaluation of solved cadmium export at the annual timescale: application to the Gironde consensus-based sediment quality guidelines for freshwater ecosystems. Arch. Environ. Contam. Toxicol. 39, 20e31. Estuary. Estuar. Coast. Shelf. S. 84, 429e 439. Milliman, J.D., Meade, R.H., 1983. World-wide delivery of river sediment to the Dalai, T.K., Nishimura, K., Nozaki, Y., 2 005 . Geochemistry of molybdenum in the oceans. J. Geol. 91, 1e21. Chao Phraya River Estuary, Thailand: role of suboxic diagenesis and porewater Milliman, J.D., Ren, M.E., 1995. River Flux to the Sea, Impact of Human Intervention transport. Chem. Geol. 218, 189e 20 2. on River Systems and Adjacent Coastal Areas. Impact on Coastal Habitation. CRC Datta, D.K., Subramanian, V., 1998. Distribution and fractionation of heavy metals in Press, pp. 57e 83. the surface sediments of the Ganges-Brahmaputra-Meghna river system in the Millward, G.E., Liu, Y.P., 2003. Modelling metal desorption kinetics in estuaries. Sci. Bengal basin. Environ. Geol. 36, 93e101. Total. Environ. 314, 613e623. de Souza Machado, A.A., Spencer, K., Kloas, W., Toffolon, M., Zarfl, C., 2016. Metal Minh, L.Q., Tuong, T.P., van Mensvoort, M.E.F., Bouma, J., 1997. Contamination of fate and effects in estuaries: a review and conceptual model for better under- surface water as affected by land use in acid sulfate soils in the Mekong River standing of toxicity. Sci. Total. Environ. 541, 268e 28 1. Delta, Vietnam. Agricul. Ecosyst. Environ. 61, 19e27. Du Laing, G., De Vos, R., Vandescasteele, B., Lesage, E., Tack, F.M.G., Verloo, M.G., Müller, G., 1979. Schwermetalle in den Sedimenten des Rheins e Vera€nderungen 20 08 . Effect of salinity on heavy metal mobility and availability in intertidal e sediments of the Scheldt estuary. Estuar. Coast. Shelf S 77, 589e 602. seit 1971. Umschau 79, 778 783. Murphy, J., Riley, J.P., 1962. A modified single solution method for the determination Etcheber, H., Taillez, A., Abril, G., Garnier, J., Servais, P., Moatar, F., Commarieu, M.V., of phosphate in natural waters. Anal. Chim. Acta 27, 31e36. 2007. Particulate organic carbon in the estuarine turbidity maxima of the Nguyen, V.L., Ta, T.K.O., Tateishi, M., 2 0 0 0 . Late Holocene depositional environments Gironde, Loire and Seine estuaries: origin and lability. Hydrobiology 588, and coastal evolution of the Mekong River Delta, Southern Vietnam. J. Asian 245e 259. Earth Sci. 18, 27e 39. Fo€rstner, U., Wittmann, G.T.W., 1979. Metal Pollution in the Aquatic Environment. Noh, S., Choi, M., Kim, E., Nguyen, P.D., Bui, X.T., Nguyen, T.V.H., Sthiannopkao, S., 486 S., 102 Abb., 94 Tab. Berlin-Heidelberg-New York 1979. Springer-Verlag. Han, S., 2013. Influence of salinity intrusion on the speciation and partitioning Gaillardet, J., Dupre´, B., Alle`gre, C.J., 1999. Geochemistry of large river suspended of mercury in the Mekong River Delta. Geochim. Cosmochim. Acta 106, sediments: silicate weathering or crustal recycling? Geochim. Cosmochim. Acta 379e 390. 63, 4037e4051. Nozaki, Y., 2001. Elemental distribution: overview. In: Steele, J.H., Thorpe, S.A., Gaillardet, J., Viers, J., Dupre´,B., 2014. Trace Elements in river waters. In: Turekian, K.K. (Eds.), Encyclopedia of Ocean Sciences, vol. 2. Academic Press, Holland, H.D., Turekian, K.K. (Eds.), Treatise on Geochemistry, second ed., vol. 7. London, pp. 840e 845. Oxford Elsevier, pp. 195e 2 35. Paikaray, S., Banerjee, S., Mukherji, S., 2 0 0 5 . Sorption of arsenic onto Vindhyan Garnier, J., Billen, G., Ne´mery, J., Sebilo, M., 2010. Transformations of nutrients (N, P, shales: role of pyrite and organic carbon. Curr. Sci. India 10, 1580e1585. Si) in the turbidity maximum zone of the Seine estuary and export to the sea. Redman, A.D., Macalady, D.L., Ahmann, D., 2 0 0 2. Natural organic matter affects Est. Coast. Shelf S 90, 129e141. arsenic speciation and sorption onto hematite. Environ. Sci. Technol. 36, Garnier, J., Lassaletta, L., Billen, G., Romero, E., Grizzetti, B., Ne´mery, J., Le, T.P.Q., 2889e 2 89 6. Wang, H., Wang, J., Liu, R., Yu, W., Shen, Z., 2015. Spatial variation, environmental Renaud, F.G., Künzer, C., 2012. Introduction. In: Renaud, F.G., Künzer, C. (Eds.), The risk and biological hazard assessment of heavy metals in surface sediments of Mekong Delta System: Interdisciplinary Analyses of a River Delta, pp. 3 e6 . the Yangtze River estuary. Mar. Pollut. Bull. 15, 250e 258. Springer Environ Sci Engineer. Wang, D., Xia, W., Lu, S., Wang, G., Liu, Q., Moore, W.S., Chen, C.T.A., 2016a. The Schultz, M.F., Benjamin, M.M., Ferguson, J.F., 1987. Adsorption and desorption of nonconservative property of dissolved molybdenum in the western Taiwan metals on ferrihydrite: reversibility of the reaction and sorption properties of Strait: relevance of submarine groundwater discharges and biological utiliza- the regenerated solid. Environ. Sci. Technol. 21, 863e 8 69 . tion. Geochem. Geophys. Geosyst 17, 28e 43. Smith, K.S. 1999. Metal sorption on mineral surfaces: an overview with examples Wang, Y., Liu, R.H., Zhang, Y.Q., Cui, X.Q., Tang, A.K., Zhang, L.J., 2016b. Transport of relating to mineral deposits. In reviews in Economic geology Vol 6A, the heavy metals in the Huanghe River estuary, China. Environ. Earth Sci. 75, 288. environmental geochemistry of mineral deposits, Eds Society of Economic Wang, Y., Ling, M., Liu, R.H., Yu, P., Tang, A.K.K., Luo, X.X., Ma, Q., 2017. Distribution Geologists Inc. (ISSN 0741e0123). and source identification of trace metals in the sediment of Yellow River Es- tuary Statham, P.J., 2012. Nutrients in estuaries- an overview and the potential impacts of and the adjacent Laizhou Bay. Phys. Chem. Earth. http://dx.doi.org/ climate change. Sci. Total. Environ. 434, 213e227. 10.1016/j.pce.2017.02.002. € Strady, E., Blanc, G., Scha fer, J., Coynel, A., Dabrin, A., 2009. Dissolved uranium, Wedepohl, H., 1995. The composition of the continental crust. Geochim.Cosmochim. vanadium and molybdenum behaviours during contrasting freshwater dis- Acta 59, 1217e1232. charges in the Gironde Estuary (SW France). Estuar. Coast. Shelf S 83, 550e 56 0 . Wen, L.S., Jiann, K.T., Santschi, P.H., 2 00 6 . Physicochemical speciation of bioactive Strady, E., Dang, V.B.H., Ne´mery, J., Gue´dron, S., Dinh, Q.T., Denis, H., Nguyen, P.D., trace metals (Cd, Cu, Fe, Ni) in the oligotrophic South China Sea. Mar. Chem. 101, 2017. Baseline seasonal investigation of nutrients and trace metals in surface 104e129. waters and sediments along the Saigon River basin impacted by the megacity of Wilbers, G.J., Becker, M., Nga, L.T., Sebesvari, Z., Renaud, F.G., 2014. Spatial and Ho Chi Minh (Vietnam). Environ. Sci. Poll. Res. 24, 3226e 3 24 3. temporal variability of surface water pollution in the Mekong Delta, Vietnam. € Toan, P.V., Sebesvari, Z., Bl asing, M., Rosendahl, I., Renaud, F.G., 2013. Pesticide Sci. Total. Environ. 485, 653e 6 65 . management and their residues in sediments and surface and drinking water in Windom, H., Smith, R., Niencheski, F., Alexander, C., 2 000 . Uranium in rivers and e the Mekong Delta. Vietnam. Sci. Total Environ. 452, 28 39. estuaries of globally diverse smaller watersheds. Mar. Chem. 68, 307e321. Trinh, A.D., Meysman, F., Rochelle-Newall, E., Bonnet, M.P., 2012. Quantification of World Bank, 2013. Vietnam Urban Wastewater Review Executive Summary. http:// sediment-water interactions in a polluted tropical river through biogeochem- www.worldbank.org/content/dam/Worldbank/document/EAP/Vietnam/vn- ical modeling. Glob. Biogeochem. Cycles 26, GB3010. urbanwastewater-summary-EN-final.pdf. Turner, A., Millward, G.E., Bale, A.J., Morris, A.W., 1993. Application of the Kd Xue, Z., Liu, J.P., Ge, Q., 2011. Changes in hydrology and sediment delivery of the concept to the study of trace metal removal and desorption during estuarine Mekong River in the last 50 years: Connection to damming, monsoon, and mixing. Estuar. Coast. Shelf S 36, 1e13. ENSO. Earth Surf. Process. Landforms 36, 296e 3 08 . Viers, J., Dupre´, B., Gaillardet, J., 2 009 . Chemical composition of suspended sedi- Xue, Z., He, R., Liu, J.P., Warner, J.C., 2012. Modeling transport and deposition of the ments in World Rivers: new insights from a new database. Sci. Total. Environ. Mekong River sediment. Cont. Shelf Res. 37, 66e 78. 407, 853e 868. Yang, Z., Xia, X., Wang, Y., Ji, J., Wang, D., Hou, Q., Yu, T., 2014. Dissolved and par- Waibel, G., Benedikter, S., Reis, N., Genschick, S., Nguyen, L., Huu, P.C., Be, T.T., 20 12. ticulate partitioning of trace elements and their spatialetemporal distribution Water governance under renovation? Concepts and practices of IWRM in the in the Changjiang River. J. Geochem. Explor 145, 114e123. Mekong Delta, Vietnam. In: Renaud, F.G., Künzer, C. (Eds.), The Mekong Delta Zhang, J., Liu, C.L., 2 00 2 . Riverine composition and estuarine geochemistry of par- System: Interdisciplinary Analyses of a River Delta, pp. 167e20 0. Springer En- ticulate metals in China: weathering features, anthropogenic impact and viron Sci Engineer, (Chapter 6). chemical fluxes. Estuar. Coast. Shelf S. 54, 1051e1070. Wang, D., Lin, W.F., Yang, X.Q., Zhai, W.D., Dai, M.H., Chen, C.T.A., 2012a. Occurrences Zhang, D., Zhang, X., Tian, L., Ye, F., Huang, X., Zeng, Y., Fan, M., 2013. Seasonal and of dissolved trace metals (Cu, Cd, and Mn) in the Pearl River Estuary (China), a spatial dynamics of trace elements in water and sediment from Pearl River large river-groundwater-estuary system. Cont. Shelf Res. 50, 54e 63. Estuary, South China. Environ. Earth Sci. 68, 1053e1063. Wang, C., Liu, S., Zhao, Q., Deng, L., Dong, S., 2012b. Spatial variation and contam- Zhao, S., Feng, C.H., Wang, D.X., Liu, Y.Z., Shen, Z.Y., 2013. Salinity increases the ination assessment of heavy metals in sediments in the Manwan Reservoir, mobility of Cd, Cu, Mn, and Pb in the sediments of Yangtze Estuary: relative role Lancang River. Ecotox. Environ. Saf. 82, 32e 39. of sediments’ properties and metal speciation. Chemosphere 91, 977e 984.